SaaS · Scheduling
Mya Team
Bookings, staff, and client data in one place.
- Role
- Full-Stack Developer
- Period
- 2022 to 2023
- Links
- Website ↗
Outcomes
The problem
Service businesses were relying on separate tools for scheduling, availability, and client tracking. That fragmentation led to double bookings, inconsistent data, and constant manual coordination.
The approach
Centralized the scheduling logic and enforced consistency in the back end instead of relying on manual fixes. Built the APIs and data models for appointments, staff roles, and availability updates so the system stays consistent when multiple actions happen at once, including the edge cases around booking conflicts, with real-time updates over Socket.IO and background jobs on Celery with Redis as the broker. Semantic matching runs on OpenAI embeddings with LangGraph agents for the AI integrations. On the front end, built clear views for managing schedules and client data, keeping the interface simple while the system handles the complexity underneath.
Stack
- Next.js
- Node.js
- Prisma
- PostgreSQL
- Tailwind CSS
- Shadcn UI
- LangGraph
- OpenAI Embeddings
- Celery
- Redis
- Socket.IO
- Stripe
- AWS
- Microservices
